Temple University

QUALITY ASSESSMENT OF SEARCH TERMSIN SPOKEN TERM DETECTION

Amir Harati and Joseph Picone, PhDDepartment of Electrical and Computer Engineering

Temple University

URL:

Abstract

• Spoken term detection is an extension of text-based searching that allows users to type keywords and search audio files containing spoken language for their existence.

• Performance is dependent on many external factors such as the acoustic channel, language and the confusability of the search term.

• Unlike text-based searches, the quality of the search term plays a significant role in the overall perception of the usability of the system.

• In this presentation we will review conventional approaches to keyword search.

• Goal: Develop a tool similar to the way password checking tools currently work.

• Approach: develop models that predict the quality of a search term based on its spelling (and underlying phonetic context).

Demo

Available at:

http://www.isip.piconepress.com/projects/ks_prediction/demo/current/

Methods

Acoustic distance algorithm. Phonetic distance algorithm. Feature based algorithm.

Feature based methods

Using different pattern recognition methods. Different features and different feature selection

methods.

Example results:Results for Neural Network

Machine feature set MSE(train) R(train) R2(train) MSE(eval) R(eval) R2(eval)1 All 0.014 0.724 0.526 0.017 0.624 0.3892 set1 0.012 0.753 0.568 0.015 0.692 0.4793 set2 0.013 0.735 0.541 0.015 0.686 0.474 set3 0.015 0.697 0.486 0.015 0.691 0.4775 set4 0.015 0.697 0.486 0.015 0.689 0.4756 set5 0.016 0.674 0.455 0.016 0.669 0.4487 set6 0.016 0.674 0.455 0.016 0.669 0.4488 set7 0.013 0.734 0.54 0.016 0.675 0.4559* All 0.017 0.742 0.551 0.016 0.67 0.4510* set1 0.016 0.752 0.566 0.014 0.705 0.497